Thermomechanical evolution of the crust during convergence...

Thermomechanical evolution of the crust during convergence

and deep crustal pluton emplacement in the Western

Province of Fiordland, New Zealand

Nathan R. Daczko

School of Geosciences, University of Sydney, NSW, Australia

Keith A. Klepeis

Department of Geology, University of Vermont, Burlington, Vermont, USA

Geoffrey L. Clarke

School of Geosciences, University of Sydney, NSW, Australia

Received 17 January 2001; revised 1 October 2001; accepted 14 December 2001; published 12 July 2002.

[1] Fiordland, New Zealand, contains exposures ofhigh-pressure (P = 12–14 kbar) granulite facies rocksthat form one of Earth’s largest exposed lower crustalroots of an Early Cretaceous magmatic arc. Theseexposures allowed us to examine the mechanismsand processes that controlled crustal thickening andlarge vertical displacements at the deepest levels of adeforming arc system. We present structural andmetamorphic data that show how the root of this arcwas tectonically thickened by imbricate, granulitefacies thrust zones during and after the emplacementof sheeted plutons into the middle and lower crust.The imbricate thrust zones form part of a well-exposed, 12 km wide, two-sided fold-thrust belt thatpreferentially developed in crust that was thermallysoftened by magmatism. Changes in metamorphicmineral assemblages and microstructural data fromthe contact aureole of a composite arc batholithcalled the Western Fiordland Orthogneiss recordthrust and pluton emplacement conditions of P �7–9 kbar (paleodepths of 25–30 km). These datashow that thrust imbrication and tectonic loading at30 km depth is a viable mechanism of large verticaldisplacements and crustal thickening at the deepestlevels of magmatic arcs. This mechanism produced acharacteristic up-pressure metamorphic history that issimilar to that observed in many other large magmaticbelts worldwide. INDEX TERMS: 8015 Structural Geology:

Local crustal structure; 8025 Structural Geology: Mesoscopic

fabrics; 8005 Structural Geology: Folds and folding; 8102

Tectonophysics: Continental contractional orogenic belts; 8035

Structural Geology: Pluton emplacement; KEYWORDS: New

Zealand, convergence, arc magmatism, lower crust, thrust tectonics

1. Introduction

[2] Investigations of the exposed lower crustal roots ofmagmatic arcs are important for interpreting orogenic pro-cesses in arc settings and for understanding the environ-ments where new continental crust is generated. However,many studies of tectonic processes in the deep crust arehindered by limited exposure of the lower crustal roots ofarc systems [e.g., Miller et al., 1993]. Two unresolvedissues include the origin of up-pressure metamorphism inthe deep crust of many arcs and the possible roles ofcontraction and thrust faulting during the thickening ofarc-derived crust.[3] Many previous investigations of arcs have shown that

pluton emplacement occurs synchronously with regionaltectonic activity and that this activity can involve contrac-tion, extension, or different combinations of these and othertectonic styles. In the Coast Mountains Batholith of westernNorth America, for example, some studies of arc tectonicsemphasize extensional processes [Klepeis and Crawford,1999; Crawford et al., 1999], whereas others emphasizecontraction [Ingram and Hutton, 1994; Andronicos et al.,1999]. Other studies show that the tectonic environment inarcs can change periodically involving both extension andcontraction (or transtension and transpression) at differentstages [e.g., Grocott et al., 1994; Tobisch et al., 1995]. Insome settings, displacements along ductile faults mayinfluence pluton emplacement, especially in the deep crust[Klepeis and Crawford, 1999; Crawford et al., 1999]. Inother settings, buoyancy forces rather than regional stressesmay control the ascent of magma [Miller and Paterson,1999]. This diversity of process has created confusion aboutthe roles of regional tectonic setting, thrust faulting, tectonicburial, and the interplay between thermal and deformationalprocesses during arc evolution. The role of thrust faulting atthe deepest levels of the Andean arc, for example, isuncertain because of lack of exposure. We address the rolesof these processes at the deep levels of arc systems using awell-exposed batholith in Fiordland, New Zealand.[4] Fiordland, New Zealand, contains Earth’s largest

(10,000 km2) and best-exposed example of a young (Early

TECTONICS, VOL. 21, NO. 4, 10.1029/2001TC001282, 2002

Copyright 2002 by the American Geophysical Union.0278-7407/02/2001TC001282$12.00

4 - 1

Cretaceous) granulite facies lower crustal root of a thick-ened (at least 45 km thick) magmatic arc [Gibson et al.,1988; Bradshaw, 1990; Oliver, 1990; Brown, 1996]. How-ever, despite excellent exposure and nearly three decades ofstudy, interpretations of the tectonic environment in whichthe high-pressure, granulite facies rocks formed and theprocesses that affected them are controversial. Much of thiscontroversy arises from conflicting interpretations of theorigin, age, and significance of rock structures and a wide

variety of metamorphic mineral assemblages within westernFiordland.[5] One major source of discord among workers in

Fiordland concerns the age of metamorphic assemblagesand mineral zoning patterns in the high-grade rocks thatindicate loading of the terrain [e.g., Bradshaw, 1989a;Clarke et al., 2000]. This controversy is similar to one inthe northern Cascades of the northwestern United States,where the origin and significance of up-pressure metamor-

Figure 1. (a) Geological map of Fiordland showing major lithologic divisions (after Bradshaw [1990]).Inset shows pre-Cenozoic configuration of the South Island, which places the Westland/Nelson regionadjacent to northern Fiordland (after Hill [1995a]). (b) Enlargement of George and Caswell Soundsshowing the localities of key sites.

Figure 2. (opposite) (a) Structural map of Caswell Sound showing major lithologic divisions and lines of cross sections.Thrust symbols on lines represent thrust zones that may be up to tens of meters wide. (b) Cross section (A–B) constructedfor the southern shore of Caswell Sound showing S0/S1 foliation trajectories within the zone of polyphase folding andductile shear. See text for discussion. (c) Cross section (A–C) constructed for the southern shore of Caswell Sound showingregional-scale variations in the orientation of foliations for the Caswell Sound fold-thrust belt. The section has been dividedinto four zones on the basis of structure. Inclined equals sign over a foliation trajectory indicates orthogneiss. Figures 2d–2g are lower hemisphere equal-area stereoplots of structural data from Caswell Sound. (d) Pre-fold-thrust belt structuralelements (L1CG and poles to S1CG) from the Caswell Gneiss west of the zone of imbricate thrust faults. (e) Pre-fold-thrustbelt structural elements (L1WFO and poles to S1WFO) from the Western Fiordland Orthogneiss east of the zone of imbricatethrust faults. (f ) D2CG structural elements (L2CG and poles to S2CG) from the Caswell Gneiss in the zones of polyphasefolding and ductile shear and upright folding. (g) D2ITF structural elements (L2ITF and poles to S2ITF) within the zone ofimbricate thrust faults. Abbreviations are as follows: CG, Caswell Gneiss; ITF, imbricate thrust faults; WFO, WesternFiordland Orthogneiss. See text for discussion.

4 - 2 DACZKO ET AL.: CRUSTAL EVOLUTION DURING CONVERGENCE

DACZKO ET AL.: CRUSTAL EVOLUTION DURING CONVERGENCE 4 - 3

phism is uncertain [Miller et al., 2000]. High-pressure(where pressure (P) is 12–14 kbar) granulite facies assemb-lages are reported for Paleozoic paragneiss and the Creta-ceous Western Fiordland Orthogneiss batholith, indicatingthat all units in Fiordland experienced a high-pressure event.However, the literature remains unclear on whether the arcbatholith was emplaced into the middle to deep crust (P =6–7 kbar) and loaded to the high-pressure metamorphicconditions during Cretaceous convergence [Bradshaw,1989a; Brown, 1996] or whether the up-pressure metamor-phic history is Paleozoic and the batholith was emplacedinto the deep crust during extension [Gibson et al., 1988;Gibson and Ireland, 1995].[6] We have discovered field relationships at Caswell

Sound that define a zone of intense folding and thrustingaffecting the contact aureole of the Western FiordlandOrthogneiss arc batholith (Figures 1 and 2). Althoughpreviously hypothesized [Oliver and Coggen, 1979; Brad-shaw, 1990], this find represents the first discovery of fold-thrust belt-type relationships within the western Fiordlandarc rocks. It is especially significant because it allows us totest previously proposed models describing the significanceof rock fabrics and the overall tectonic history of theCretaceous arc batholith and country rocks that containhigh-pressure (P = 12–14 kbar) garnet granulite faciesassemblages (see Clarke et al. [2000] for a review). Weshow the importance of contractional tectonics on arcevolution, including burial, high-pressure metamorphism,and the P-T-t-D path of the terrain.[7] In this paper we focus on relationships between

deformation and metamorphism that followed the emplace-ment of the Western Fiordland Orthogneiss at CaswellSound and discuss the significance of these relationshipsfor interpreting the evolution of Fiordland’s high-pressuregranulite facies belt. We compare and contrast the fold-thrust belt setting with an undeformed intrusive contact atGeorge Sound (Figure 1). We describe evidence of regionalconvergence and thrusting of Paleozoic paragneiss over theWestern Fiordland Orthogneiss batholith at middle to lowercrustal depths (�25–30 km). We conclude that the high-pressure granulite facies conditions were attained simulta-neously with and at least in part due to tectonic loading, bythe stacking of imbricated thrust sheets, during the EarlyCretaceous. These new data show that models involvinggranulite emplacement wholly by Cretaceous extension ormagmatic loading of this region are incomplete. Finally, wediscuss the wider tectonic implications of this study toconvergent margin tectonics, arc batholith emplacement,and associated contractional tectonics in general. We assessthe potential significance of this research for interpretingprocesses affecting the deep levels of magmatic arcs.

2. Regional Geologic Setting

[8] The geology of the south island of New Zealand isdivided into Eastern and Western provinces (inset Figure 1a)[Landis and Coombs, 1967; Bishop et al., 1985]. A belt ofrocks that is referred to as the Median Tectonic Zone (MTZ;inset Figure 1a) [Kimbrough et al., 1993, 1994] or the

Median Batholith [Mortimer et al., 1999] separate these twoprovinces. Most of the Eastern Province formed within aconvergent margin setting and contains arc-volcanic rocks,arc-derived sedimentary sequences, and accretionary com-plexes of Permian-Cretaceous age [MacKinnon, 1983;Bradshaw, 1989; Mortimer, 1995]. The Western Provincecontains extensive lower Paleozoic paragneiss (includingthe Caswell Gneiss of this study), cut by Devonian andCarboniferous granitoids [Muir et al., 1996; Wandres et al.,1998; Ireland and Gibson, 1998]. Rocks within this prov-ince preserve a polyphase mid-Paleozoic history thatincludes low-pressure/high-temperature metamorphism fol-lowed by medium-pressure/high-temperature metamor-phism. Paleozoic events occurred when ancestral NewZealand lay within or outboard of the Pacific margin ofGondwana [Wood, 1972; Carter et al., 1974; Gibson andIreland, 1996; Ireland and Gibson, 1998; Mortimer et al.,1999].[9] The Median Tectonic Zone is a comparatively narrow

belt of tectonically disrupted arc-related rocks with U-Pbzircon ages that mostly fall into two age groups: 247–195 Ma and 157–131 Ma (Figure 1) [Bradshaw, 1993;Kimbrough et al., 1993, 1994]. Late Triassic MedianTectonic Zone plutons that intrude the Eastern Provinceindicate that this province and the tectonic zone weretogether at this time [Williams and Harper, 1978; Mortimeret al., 1999]. Rocks of the Median Tectonic Zone and theWestern Province are intruded by stitching plutons of theEarly Cretaceous Western Fiordland Orthogneiss/Separa-tion Point Suite (Figure 1) [Bradshaw, 1990; Kimbroughet al., 1994].[10] The Western Fiordland Orthogneiss (WFO; Figure 1)

[Bradshaw, 1990] is a batholithic-scale unit of mafic tointermediate composition that was emplaced into thickenedarc crust. Maximum pressures for Fiordland crust indicatethat the arc was at least 45 km thick during the EarlyCretaceous, and geophysical studies of the present-daycrustal structure suggest that at least a further 10 km ofcrust still lies beneath Fiordland [Oliver, 1990]. Conven-tional and sensitive high-resolution ion microprobe(SHRIMP) U-Pb zircon ages range between 126 and116 Ma [Mattinson et al., 1986; Gibson et al., 1988; Gibsonand Ireland, 1995; Muir et al., 1998]. Amphibolite faciesshear zones juxtapose the Western Fiordland Orthogneisswith Paleozoic paragneiss to the northwest, west, and south(Figure 1) [Gibson et al., 1988; Hill, 1995a, 1995b; Klepeiset al., 1999]. George Sound and Mt Daniel (Figure 1)preserve relationships where the Western Fiordland Orthog-neiss intrudes Paleozoic paragneiss and the Arthur RiverComplex orthogneiss, respectively [Bradshaw, 1990]. Con-tact relationships with rocks of the Eastern Province aremostly intrusive but are poorly understood [Bradshaw,1990].[11] Ireland and Gibson [1998] observed changes in

metamorphic assemblages and used microprobe U-Pb dat-ing of zircon/monazite to infer loading of Paleozoic para-gneiss from P = 3–5 kbar to P = 7–9 kbar between 360 Maand 330 Ma. These authors also examined Cretaceous shearzones in the Doubtful Sound region (Figure 1) and inferred


the emplacement of the Western Fiordland Orthogneiss,during regional extension, into rocks that were already atlower crustal conditions (P > 12 kbar).[12] In contrast, Bradshaw [1989b, 1990] and Bradshaw

and Kimbrough [1989] used conventional U-Pb zircondating [Mattinson et al., 1986] and estimates of metamor-phic P-T paths to infer the midcrustal emplacement of theWestern Fiordland Orthogneiss batholith coeval with low-to medium-pressure (P = 6–7 kbar) granulite facies meta-morphism. A substantial increase in pressure (to P = 12–13 kbar) and the subsequent formation of garnet granulitethroughout Fiordland was attributed by these authors totectonic burial consequent to regional convergence (possi-bly involving arc-continent collision). In contrast, usingpetrographic analyses and field data reported by Bradshaw[1989a, 1989b, 1990], Oliver [1990], and Brown [1996]inferred that the high-pressure granulite facies conditionswere produced by magma loading following emplacementof the Western Fiordland Orthogneiss at midcrustal con-ditions. Finally, Muir et al. [1995, 1998] used geochemicaland geochronologic data to argue that an Early Cretaceousmagmatic arc, chemically equivalent to parts of theMedian Tectonic Zone, was thrust beneath western Fiord-land during convergence to depths in excess of 40 km andmelted to produce the Western Fiordland Orthogneiss. It istherefore unclear within the literature whether a domi-nantly convergent or extensional tectonic setting accom-panied formation of the Fiordland high-pressure granulitesand what caused the loading of these rocks. The exhuma-tion of parts of the Fiordland high-pressure metamorphicbelt has been attributed to latest Cretaceous crustal exten-sion, prior to the opening of the Tasman Sea Basin[Gibson et al., 1988], supplemented by early to lateCenozoic transpressive motion in northern Fiordland[Blattner, 1991; Klepeis et al., 1999].

3. Rock Units

[13] From east to west, Caswell Sound contains threemain rock types: (1) Western Fiordland Orthogneiss, (2)Caswell Gneiss, and (3) McKerr monzodiorite (Figure 2a).The Western Fiordland Orthogneiss is most commonlyweakly deformed hornblende-plagioclase-dominated gab-broic gneiss. However, two-pyroxene gabbroic gneiss andK-feldspar-quartz monzogranite varieties occur at somelocalities. Tabular plagioclase laths sharing long, straightgrain boundaries with pyroxene or amphibole are relictigneous textures.[14] The Caswell Gneiss consists of 60% paragneiss and

40% biotite dioritic orthogneiss. The paragneisses are bestexposed along the shores of Caswell Sound at its westernend near the contact with the McKerr monzodiorite(Figure 2a). The paragneiss comprises calc-silicate andmafic gneiss, marble, and metapsammitic schist. All unitsare interbedded, and the thickness of sedimentary layeringis highly variable ranging from a few centimeters to morethan 100 m. The biotite-plagioclase-dominated dioriticorthogneiss contains rafts of the paragneiss and is bestexposed near Walker Point (Figure 2a). The orthogneiss

forms sill-like intrusions that appear to be more than 500 mthick.[15] The McKerr monzodiorite is a very weakly deformed

biotite-plagioclase rock that cuts the Caswell Gneiss. Thecontact with Caswell Gneiss displays clasts of brecciatedparagneiss within the monzodiorite that are inferred torepresent an igneous breccia. This contact is tectonized asdiscussed in section 4.5 (Figure 2b).[16] George Sound contains Western Fiordland Orthog-

neiss that envelops and cuts rafts of intensely foliatedschists of the Paleozoic George Sound Paragneiss thatmay be up to 4 km wide (site 97-17; Figure 1b) [Bradshaw,1990]. Diatexite zones (up to 200 m wide) occur at thecontacts between the George Sound Paragneiss and theWestern Fiordland Orthogneiss (site 97-01 and 97-13;Figure 1b). The George Sound Paragneiss is most probablyequivalent to the metasedimentary units of the CaswellGneiss, though no marble horizons were observed at GeorgeSound.

4. Geometry and Fabric Elements

of the Caswell Fold-Thrust Belt

[17] Caswell Sound contains �12 km of continuousexposure across a fold-thrust belt that deforms the uppercontact of the Early Cretaceous Western Fiordland Orthog-neiss batholith (Figure 2). We divide the section into sixstructural domains. From east to west, these include (1) aweakly deformed Western Fiordland Orthogneiss, (2) a zoneof gently west dipping imbricate thrust faults that deformthe contact between the Western Fiordland Orthogneiss andCaswell Gneiss, (3) a zone characterized by upright foldsand no thrust zones, (4) a zone of polyphase folding andductile shear, (5) a zone of west vergent thrusting at thecontact between the Caswell Gneiss and the McKerr mon-zodiorite, and (6) weakly deformed to undeformed McKerrmonzodiorite. In this section we define the geometry andstructural elements of the fold-thrust belt, including themineral assemblages within each domain.

4.1. Structural Elements Deformed by the CaswellFold-Thrust Belt

[18] The Western Fiordland Orthogneiss and McKerrmonzodiorite to the east and west of the Caswell Gneiss,respectively, were weakly to very weakly deformed prior tothe development of the fold-thrust belt. The Western Fiord-land Orthogneiss displays a weakly developed foliation(S1WFO) that is most probably a reworked magmatic folia-tion. S1WFO is commonly defined by aligned and elongateclusters of mafic minerals in a dominantly feldspar matrix.Mafic mineral clusters most commonly comprise coarsecalcic-amphibole and clinozoisite with or without biotiteand rutile. In some localities, the mafic mineral clusters arecored by coarse orthopyroxene and clinopyroxene rimmedby fine amphibole. At these localities, orthopyroxene andclinopyroxene grains exhibit magmatic textures defined bycoarse, equant orthopyroxene aggregates with exsolutionblebs of clinopyroxene and opaque phases. The orientationof S1WFO is variable displaying both steep and shallow dips


(Figure 2e). On the basis of limited lineation data, a weaklydeveloped amphibole mineral lineation (L1WFO) shallowlyplunges variably to the north, northeast, and west (Figure 2e).The McKerr monzodiorite shows randomly oriented biotiteand feldspar with no evidence of a foliation or a lineation.[19] Within the Caswell Gneiss, the dominant pre-fold-

thrust belt structure is a well-developed shallowly dippingfoliation (S1CG) that subparallels bedding planes in para-gneiss and parallels the margins of sills within the biotitediorite. Foliated rafts of metasediment within the biotitediorite suggest that the paragneiss experienced a complexpre-fold-thrust belt history that, for simplicity, we omit here.The S1CG foliation is defined by lenticular aggregates ofquartz and feldspar and aligned muscovite, biotite, orchlorite grains. This foliation is axial planar to recumbent,isoclinal folds of bedding (Figure 2b). The foliation planescontain a well-developed shallowly SSW plunging musco-vite and biotite mineral lineation (Figure 2d). The axialplanar S1CG foliation and recumbent folds are folded by

upright, gently south plunging F2CG folds (Figure 2b), suchthat the early folds are no longer recumbent in many places.

4.2. Zone of Imbricate Thrust Faults (ITF)

[20] The zone of imbricate thrust faults extends west-ward from Green Point within the Western FiordlandOrthogneiss to Walker Point within the Caswell Gneiss(Figure 3a). A traverse westward along Caswell Soundfrom Green Point shows �10 shallowly west dippingoutcrop-scale thrust zones that display listric geometries.The thrust zones cut the Western Fiordland Orthogneiss,the contact aureole proximal to the Western FiordlandOrthogneiss, and the Caswell Gneiss foliation outside thecontact aureole. The thrust zones consist of narrow(<250 m wide) zones of highly deformed and recrystallizedWestern Fiordland Orthogneiss and/or Caswell Gneiss. Thespacing between fault zones is �200–400 m. Well-devel-oped foliations (S2ITF) in the thrust zones are locally

Figure 3. (a) Structural map of the zone of imbricate thrust faults. Boundaries of the zone of imbricatethrust faults are shown as fine dashed lines. Thrust symbols on lines represent thrust zones that may be upto tens of meters wide as detailed in Figure 3b. Note that thrust zones cut the Western FiordlandOrthogneiss. (b) Vertical section B-B0 shown in Figure 3a of a thrust zone in the Caswell Gneiss viewedperpendicular to transport direction. Note zones of ultramylonite and fault propagation folds.


mylonitic to ultramylonitic. S2ITF foliation planes containwell-developed mineral lineations that plunge gently to thewest (Figures 2g, 4b and 4c). The ultramylonitic bandsshow small ramp-flat geometries that include curvatures in

the fault surfaces (Figure 3b). Thrust zones are mostcommonly found in rheologically weak zones that includethe contact between Western Fiordland Orthogneiss andcountry rock and marble layers (Figure 3b). Approximately80% of the imbricate thrust zones are located within thinmarble horizons. In many cases it is unclear whether themarbles are interbeds or thrust zone slices. At least someof the marble horizons contain less-deformed boundariesand are thick enough to suggest they represent originalbedding horizons. Figure 3b shows common features ofindividual thrust zones. These include fault propagationfolds [Suppe, 1985] that accommodate fault displacementwhere ultramylonitic ramps terminate. Folds (F2ITF) ofcompositional layering, S1CG, and S1WFO between indi-vidual thrust zones are commonly rootless with axialplanes that lie parallel to the shallow to moderately westdipping mylonitic to ultramylonitic folia (Figures 3b, 4band 4c).[21] Western Fiordland Orthogneiss in contact with the

Caswell Gneiss is a monzogranite. It has the S2ITF assem-blage garnet, biotite, plagioclase, K-feldspar, and quartzwith or without muscovite, titanite, and hematite. At mostother localities the Western Fiordland Orthogneiss is dioriticto gabbroic gneiss with a similar assemblage that lacksgarnet. Thrust zones that deform Caswell Gneiss within500 m of the contact with the Western Fiordland Orthog-neiss contain the S2ITF assemblage garnet, biotite, plagio-clase, K-feldspar, and quartz in biotite diorite and psammiticschist. Thrust zones in the Caswell Gneiss more than 500 mand up to 2.5 km from the contact with the WesternFiordland Orthogneiss contain the S2ITF assemblage chlor-ite, muscovite, and quartz with porphyroclasts of plagio-clase, K-feldspar, clinozoisite, and amphibole in biotitediorite and psammitic schist.

4.3. Zone of Upright Folding

[22] For �4 km west of the imbricate thrust zones, S1CG

is deformed by open to tight, upright F2CG folds of thesheeted biotite dioritic orthogneiss (Figures 2b and 2c).These folds deform the early shallowly dipping foliation(S1CG) and sills of biotite diorite in the Caswell Gneiss. Awell-developed foliation (S2CG,; Figure 4a) formed axialplanar to the F2CG folds. The assemblage garnet, biotite,muscovite, chlorite, plagioclase, and quartz most commonlydefines S2CG in metapelitic schists. In this zone of uprightfolding, S2CG foliation planes strike north and dip steeply tothe east and west (Figures 2f and 4a). S2CG foliation planescontain a well-developed mineral lineation, defined mostlyby biotite and muscovite that plunges shallowly to thesouth, subparallel to F2CG fold axes. At the eastern boun-dary of this zone (near Walker Point), S2CG foliation planesgradually dip more gently to the west to lie parallel to theshallow to moderately west dipping S2ITF foliation planes ofthe imbricate thrust zone.

4.4. Zone of Polyphase Folding and Ductile Shear

[23] West of the zone of upright folding is a 3–4 km widezone of polyphase, tight folding, and ductile shear. The

Figure 4. (a) Outcrop photograph of a well-developed,steeply dipping S2CG foliation that is axial planar to F2CG

folds of a composite S0/S1 in the zone of polyphase foldingand ductile shear (view to south). Person for scale. (b)Outcrop photograph showing D2ITF thrust zone in theCaswell Gneiss (view to south). This outcrop was the sourcefor Figure 3b. Person for scale. (c) Outcrop photographshowing D2ITF thrust zone in the Western FiordlandOrthogneiss (view to south). Marker pen (�13 cm long)for scale.


eastern boundary of this zone is gradational from the zoneof upright folding across �500 m where folds becometighter westward. The western boundary of this zone is awest vergent thrust near Hansard Point (Figure 2a). Themost distinguishing feature of this zone is a 1.5-km-wideductile shear zone that deforms a succession of calc-silicate,metapelitic, psammitic, and marble units. The shear zone ischaracterized by upright, asymmetric, tight to isoclinal foldsand a very well-developed steeply dipping foliation (S2CG;Figures 4a and 5). This shear zone foliation crosscuts allearly shallowly dipping foliations and recumbent folds inthe Caswell Gneiss. The tight upright folds are coaxial withthe older recumbent folds and produce a macroscopic type-3interference pattern (fishhooks). With the exception of anarea of ‘‘M-type’’ symmetrical fold geometry in the centralregion of this zone, most of these folds are overturned andwest vergent (Figures 2b and 2c). Proximal to the zone ofwest vergent thrusting, the near vertical axial planes of theseF2CG folds become inclined and subparallel the westvergent thrusts (Figure 2b).

4.5. Zone of West Vergent Thrusting

[24] A zone of west vergent thrusting is defined by anarrow (<50 m) series of semi-brittle, east-dipping thrustzones that separate the most highly deformed part of thesection (the ductile shear zone described in the previoussection) from the weakly deformed McKerr monzodiorite.The narrow width of the west vergent thrust zone and thesmall number of thrusts within it suggest that this style anddirection of thrusting is subordinate to the more abundanteast vergent thrust zones in the imbricate zone.

5. Comparison of Western Fiordland

Orthogneiss Contact Relations:

Caswell Sound Versus George Sound

[25] Caswell Sound contains exposures of the upper(southwestern) contact of the Western Fiordland Orthog-neiss batholith. George Sound preserves a 4-km-wide sec-tion of metasedimentary rock that is wholly enveloped bythe Western Fiordland Orthogneiss. We infer that GeorgeSound exposes rock structurally below the upper contactpreserved at Caswell Sound. Patterns of deformationobserved at George Sound differ in style and intensity from

those described above for Caswell Sound. A well-preserveddiatexite and a weakly deformed contact zone suggest thatthe contact aureole at George Sound was shielded frommost of the effects of Early Cretaceous contractional defor-mation. The absence of thrust zones or other intensecontractional deformation at George Sound suggests thatEarly Cretaceous deformation was partitioned into the uppercontact at Caswell Sound where weak, thermally softenedmarble and calc-silicate horizons localized thrust zones atthe Western Fiordland Orthogneiss boundary.

6. Metamorphism and Conditions of

Deformation

[26] In this section we present calculated temperature andpressure conditions from the metamorphic assemblagesoutlined below. Finally, we present evidence for a 500-m-wide thermal gradient in the contact aureole of the WesternFiordland Orthogneiss at Caswell Sound using variations inmetamorphic mineral assemblage and quartz and feldsparmicrostructures.

6.1. Calculated Pressure and Temperature Conditions

[27] The S1CG mineral assemblage within the CaswellGneiss includes garnet, amphibole, biotite, plagioclase, andquartz. At George Sound foliated schists in country rockoutside the diatexite zone contain the assemblage garnet,kyanite, biotite, staurolite, plagioclase, and quartz. On thebasis of crosscutting relationships we infer that theseassemblages and the foliations they define formed prior toemplacement of the Western Fiordland Orthogneiss.[28] The high-grade thrust zones within and near the

Western Fiordland Orthogneiss contact at CaswellSound (site 99-35; Figure 1b) contain foliations definedby the metamorphic assemblage garnet, biotite, plagioclase,K-feldspar, and quartz. Thrust zones more than 500 m fromthe contact with the Western Fiordland Orthogneiss (site99-29; Figure 1b) do not contain assemblages suitable forcalculating metamorphic conditions. However, garnet-bear-ing S2CG assemblages occur in metapelitic schists in thezone of polyphase folding and ductile shear (site 97-14;Figures 1b, 2c, and 4a). These assemblages are inferred tohave formed synchronously with those in the zone ofimbricate thrust faults on the basis of a correlation ofS2CG and F2CG structures across Caswell Sound, a lackof crosscutting relationships between these structures acrossstructural domains, and similar conditions of metamor-phism (e.g., sites 97-14 and 97-29; Figure 1b).[29] We use the assemblages outlined above to estimate

metamorphic temperature and pressure paths prior to andduring development of the Caswell fold-thrust belt. We alsocalculate P-T conditions that accompanied D2CG/D2ITF andcompare these temperatures with estimates obtained usingmicrostructures. Representative microprobe analyses ofequilibrium assemblages are presented in Table 1a. We useddirectly calibrated thermometry [Ferry and Spear, 1978;Graham and Powell, 1984] and the average P-T approachusing the computer software THERMOCALC (version 2.6)[Powell and Holland, 1988] (Table 1b). For average P-T

Figure 5. Sketch of foliation and superposed foldrelationships in the zone of polyphase folding and ductileshear.


Table

1a.RepresentativeMicroprobeAnalyses(w

t%

OxideandCationData)

ofSelectedEquilibrium

Assem

blages

a

GeorgeSoundDiatexite

Sam

ple

97-01J2

S2CG(ZoneofPolyphaseFoldingandDuctileShear)

Sam

ple

99-14A

S2TZ(ZoneofIm

bricateThrustFaults-

CaswellGneiss)Sam

ple

99-35

S2ITZ(ZoneofIm

bricateThrustFaults-WFO)

Sam

ple

99-35D

gbi

amph

pl

gbi

mu

pl

epchl

gbi

pl

ksp

gbi

pl

ksp

SiO

236.95

35.23

39.29

63.19

36.68

34.22

45.06

59.47

37.98

24.72

36.77

35.64

58.27

64.91

36.84

34.69

62.06

63.95

TiO

20.06

2.47

0.73

0.01

0.08

1.66

0.65

0.02

0.07

0.09

0.01

2.52

0.00

0.03

0.03

3.12

0.02

0.00

Al 2O3

20.72

15.91

14.26

22.89

21.24

18.49

34.53

25.75

26.98

21.83

21.22

17.75

26.24

18.16

20.43

15.90

23.87

18.83

Cr 2O3

0.02

0.02

0.08

0.00

0.00

0.04

0.00

0.00

0.00

0.00

0.03

0.02

0.02

0.00

0.00

0.03

0.08

0.00

FeO

28.62

21.34

21.97

0.17

26.79

22.34

2.60

0.04

9.02

26.59

32.91

18.96

0.10

0.03

28.33

26.62

0.02

0.08

MnO

4.24

0.13

0.30

0.01

8.46

0.17

0.00

0.00

0.36

0.36

3.05

0.11

0.03

0.00

5.09

0.42

0.03

0.00

MgO

1.98

9.21

6.08

0.00

1.71

9.25

0.56

0.01

0.01

15.14

2.95

9.26

0.02

0.02

1.44

5.03

0.01

0.02

CaO

7.41

0.01

10.19

4.08

4.84

0.04

0.00

7.02

23.25

0.01

3.40

0.00

8.16

0.00

7.51

0.00

5.06

0.03

Na 2O

0.09

0.17

2.31

9.31

0.03

0.18

0.90

7.53

0.01

0.03

0.02

0.11

6.97

0.19

0.06

0.03

8.58

0.87

K2O

0.01

9.26

1.24

0.09

0.00

9.55

10.02

0.06

0.01

0.02

0.02

9.45

0.07

16.32

0.01

9.55

0.21

15.91

Total

100.09

93.76

96.44

99.74

99.82

95.95

94.32

99.90

97.68

88.79

100.38

93.83

99.87

99.64

99.73

95.38

99.93

99.69

#O

12

22

23

812

22

22

825

28

12

22

88

12

22

88

Si

3.0

5.5

6.2

2.8

3.0

5.3

6.1

2.7

6.1

5.2

3.0

5.5

2.6

3.0

3.0

5.5

2.8

3.0

Ti

0.0

0.3

0.1

0.0

0.0

0.2

0.1

0.0

0.0

0.0

0.0

0.3

0.0

0.0

0.0

0.4

0.0

0.0

Al

2.0

2.9

2.6

1.2

2.0

3.4

5.5

1.4

5.1

5.4

2.0

3.2

1.4

1.0

2.0

3.0

1.2

1.0

Cr

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

Fe

1.9

2.8

2.9

0.0

1.8

2.9

0.3

0.0

1.2

4.7

2.2

2.5

0.0

0.0

1.9

3.5

0.0

0.0

Mn

0.3

0.0

0.0

0.0

0.6

0.0

0.0

0.0

0.0

0.1

0.2

0.0

0.0

0.0

0.3

0.1

0.0

0.0

Mg

0.2

2.2

1.4

0.0

0.2

2.1

0.1

0.0

0.0

4.7

0.4

2.1

0.0

0.0

0.2

1.2

0.0

0.0

Ca

0.6

0.0

1.7

0.2

0.4

0.0

0.0

0.3

4.0

0.0

0.3

0.0

0.4

0.0

0.7

0.0

0.2

0.0

Na

0.0

0.1

0.7

0.8

0.0

0.1

0.2

0.7

0.0

0.0

0.0

0.0

0.6

0.0

0.0

0.0

0.7

0.1

K0.0

1.9

0.2

0.0

0.0

1.9

1.7

0.0

0.0

0.0

0.0

1.9

0.0

1.0

0.0

1.9

0.0

0.9

Total

8.0

15.7

15.9

5.0

8.0

15.8

14.1

5.0

16.4

20.1

8.0

15.5

5.0

5.0

8.0

15.6

5.0

5.0

aDatawereobtained

usingaCam

ecaSX50microprobehousedat

theUniversity

ofNew

South

Wales

runningat

anacceleratingvoltageof15kVandabeam

width

of1–5m.


estimates, we used the internally consistent thermodynamicdata set of Holland and Powell [1990] (data file createdApril 1996). An explanation of using this method forfinding an average P-T of the mineral assemblages ispresented by Clarke et al. [2000]. The results of directlycalibrated thermometry and average P-T calculations(assuming the activity of water was 1.0 on the basis ofhydrous assemblages) are presented in Table 1b. TheTHERMOCALC barometry results presented in Table 1bbecome 1–2 kbar lower, and the THERMOCALC ther-mometry results become 60�–80�C lower if an activity ofwater of 0.5 is assumed. Errors (2s) are presented on theaverage P and average T calculations. Only rims of minerals(most often in grain contact) were used for the thermobar-ometry.[30] Several observations suggest that the conditions of

metamorphism we have determined occurred during each ofthe events we outline. First, the minerals used in calculatingP-T estimates define the structural elements in the variousstructural domains. Second, textural relationships suggestsyntectonic mineral growth. For example, biotite and mus-covite in the schists define mica fish structures consistentwith growth during deformation. Third, the relative timingof mineral growth, especially within the Western FiordlandOrthogneiss, is well constrained. Garnet, for example, isonly observed in samples intensely deformed within thezone of imbricate thrust faults, indicating syn-D2ITF mineralgrowth. Finally, only the rims of minerals in grain contactdisplaying equilibrium textures were probed and used in thedetermination of pressure and temperature conditions.[31] Garnet-hornblende and garnet-biotite thermometry

for the S1CG assemblage yielded T = 500�–600�C for P =4 kbar. The average P-T result suggests that the S1CG

assemblage equilibrated at P = 4.2 ± 1.88 kbar and T =543� ± 64�C, which is consistent with the directly cali-brated thermometry. The average P-T result for the schistsat George Sound (site 99-17; Figure 1b), outside of thezone of diatexite, yielded estimates of P = 7.6 ± 1.78 kbarand T = 632� ± 42�C which is higher pressures andtemperatures than estimated for the S1CG assemblage.[32] The diatexite assemblages at George Sound from

site 97-01 yielded directly calibrated garnet-hornblendeand garnet-biotite thermometry estimates of T = 550�–700�C for P = 8 kbar (Table 1b). The average P-T resultsuggests that the diatexite assemblage equilibrated at P =8.7 ± 2.1 kbar and T = 644� ± 86�C (Table 1b), which iswithin error of the temperature estimate obtained using

directly calibrated thermometry. The average P-T result fora second sample of diatexite at George Sound (site 99-13;Figure 1b) yielded estimates of P = 8.6 ± 1.52 kbar andT = 726� ± 46�C.[33] Garnet-biotite thermometry for the S2CG assemblage

(site 99-14; Figure 1b) gives T = 550�C for P = 8 kbar. Theaverage P-T result suggests that the S2CG assemblageequilibrated at P = 6.9 ± 0.96 kbar and T = 551� ± 22�C,consistent with results from the directly calibrated thermom-etry. Garnet-biotite thermometry for the high-grade S2ITF

assemblages in the Caswell Gneiss and Western FiordlandOrthogneiss (site 99-35; Figure 1b) gives T = 600�–760�Cfor P = 8 kbar. The average P-T result suggests that theS2ITF assemblages equilibrated at P = 7.6 ± 3.12 kbar, T =813� ± 132�C and P = 7.9 ± 2.78 kbar, T = 794� ± 148�C forthe high-grade Caswell Gneiss and Western FiordlandOrthogneiss samples, respectively.

6.2. Fault Zone Microstructures Inside and Outsidethe Contact Aureole

[34] In the previous section we showed that mineralassemblages defining the S2CG and S2ITF foliations varyacross the structural domains at Caswell Sound. Theassemblages and calculated temperatures that accompanieddeformation in these zones suggest that the temperature washigher adjacent to the Western Fiordland Orthogneiss.Samples within the Western Fiordland Orthogneiss and upto �500 m west of its contact yielded results consistent withgranulite facies conditions. In contrast, samples within theCaswell Gneiss, deformed by the fold-thrust belt more than500 m distal to the contact with the Western FiordlandOrthogneiss, give results consistent with amphibolite faciesconditions (compare temperatures of 700�–800�C withinthe contact aureole to estimates of 550�–600�C outside it).In this section, we describe variations in microstructurewithin the zone of imbricate thrust faults that support ourinterpretation of a thermal gradient.[35] To better define the effects of a thermal gradient

within the contact aureole, we examined and compareddeformation textures in three domains within the zone ofimbricate thrust faults: (1) the easternmost thrust zones thatcut the Western Fiordland Orthogneiss batholith (e.g., site99-35; Figure 1b); (2) a thrust zone that deformed theCaswell Gneiss within 100 m of the contact with theWestern Fiordland Orthogneiss; and (3) the westernmostimbricate thrust zones that deformed the Caswell Gneiss

Table 1b. Directly Calibrated Thermometry, Average P-T Calculations of Selected Equilibrium Assemblagesa

Method

George SoundDiatexite Sample

97-01J2

S2CG (Zone of PolyphaseFoldling and Ductile Shear)

Sample 99-14A

S2TZ (Zone of ImbricateThrust Faults-CaswellGneiss) Sample 99-35

S2ITZ (Zone of ImbricateThrust Faults-WFO)Sample 99-35D

1, 2 5561, 7032 �C 5431 �C 6011 �C 759 �C3 T = 644� ± 86�C,

P = 8.7 ± 2.1 kbarT = 551� ± 22�C,P = 6.9 ± 0.96 kbar

T = 813� ± 132�C,P = 7.6 ± 3.12 kbar

T = 794� ± 148�C,P = 7.9 ± 2.78 kbar

aThermobarometry methods are (1) garnet-biotite thermometry [Ferry and Spear, 1978], (2) garnet-hornblende thermometry [Grahamand Powell, 1984], and (3) average P-T approach of THERMOCALC.


2.5 km west of the contact (e.g., site 99-29; Figure 1b). Foreach domain, samples were collected across the zones andincluded mylonitic to ultramylonitic samples.[36] Thrust zones within the Western Fiordland Orthog-

neiss occur up to a few hundred meters east of its contactwith the Caswell Gneiss (Figure 3a). Samples of WesternFiordland Orthogneiss unaffected by thrust zones display aweakly developed foliation defined by amphibole or two-pyroxene assemblages. These samples show limited evi-dence of deformation during D2ITF such as warped twins inplagioclase grains which otherwise generally display agranoblastic texture. Undulose extinction in quartz and benttwins in plagioclase are common.[37] Samples of Western Fiordland Orthogneiss from

thrust zones contain a well-developed biotite-rich S2ITF

foliation with dynamic recrystallization of both feldsparand quartz grains. Large (0.5 mm) quartz grains exhibitstrong subgrain development, rotational recrystallizationand grain boundary migration. Elongate feldspar grains

define the foliation and display core and mantle texturesconsisting of serrated boundaries surrounded by a mantle offine recrystallized grains. It is common for the boundarybetween core and mantle to be blurred or poorly defined(Figure 6a). Feldspar cores show some undulatory extinc-tion, but no fracturing of grains is evident. K-feldspardisplays synthetic twinning along the boundaries of inclu-sions and some grains are microperthitic. Some K-feldspargrains show flame-shaped albite lamellae (Figure 6b) thatoriginate at grain boundaries, typically in pressure shadowpositions. Flames are generally widest at the grain boundaryand taper to a point within the grain (Figure 6b). Thesesamples also commonly contain narrow bands (<1 mm) ofultramylonite.[38] Mylonitic to ultramylonitic samples from thrust

zones within the Western Fiordland Orthogneiss containlarge (0.5 cm) feldspar porphyroclasts displaying unduloseextinction that also are mantled by dynamically recrystal-lized quartz and feldspar grains. The proportion of dynam-

Figure 6. Photomicrographs of mylonite textures in samples from across the Caswell Sound fold-thrustbelt. Width of view for each photomicrograph is 3.9 mm. All photomicrographs are taken in crossedpolarized light. Figures 6a–6c are samples of Western Fiordland Orthogneiss from site 99-35. Figures6d–6f are samples of Caswell Gneiss cut by the thrust zone at site 99-29, shown in Figure 6b. (a)Dynamically recrystallized quartz and feldspar. The feldspar porphyroclast at top (arrow) is mantled byrecrystallized feldspar. Note boundaries of porphyroclast are serrated. (b) K-feldspar grain with perthiticflames (arrow). (c) Quartz subgrains and bent plagioclase twins (arrow). (d) S-C0 fabric showing top tothe right (east) sense of shear. S planes run from left to right, C0 planes run from top left to bottom right.Note zone of ultramylonite at top of photograph. (e) Antithetic microfault in a feldspar porphyroclast.Arrow indicates direction of displacement for the upper side of the fault (i.e., top down to the left (west)sense of shear). (f ) Synthetic microfault in a clinozoisite porphyroclast. Arrow points in direction ofdisplacement for the upper side of the fault (i.e., top to the right (east) sense of shear).


ically recrystallized material is much greater in the mylo-nitic to ultramylonitic samples compared to those describedabove. Polycrystalline quartz ribbons surround the feldsparporphyroclasts. Many of the quartz ribbons contain elongatesingle crystals that lack evidence of intracrystalline defor-mation structures. Ribbons of feldspar are also found inthese samples. Feldspar porphyroclasts show well-devel-oped asymmetric tails of recrystallized matrix material.Myrmekite is occasionally found along foliation planes.[39] Within the Caswell Gneiss samples from thrust

zones situated <500 m from the contact show abundantevidence of ductile deformation similar to that describedabove for the Western Fiordland Orthogneiss. These rocksdisplay evidence of the dynamic recrystallization of feldsparand quartz and recrystallization through the nucleation andgrowth of new grains at grain boundaries or in segregationbands. These microstructural observations combined withthose of Olsen and Kohlstedt [1985] and Pryer [1993]suggest that the temperatures during deformation in thispluton/contact aureole zone was >550�C.[40] In contrast, thrust zones more than 500 m from the

contact show different textures. For example, samples 2 kmwest of the Caswell Gneiss/Western Fiordland Orthogneisscontact display little evidence of dynamically recrystallizedfeldspar. Evidence of subgrain development in quartzincludes grains that are mantled by other minor finer-

grained dynamically recrystallized quartz grains. Plagio-clase porphyroclasts commonly display bent twins, frac-tures, and undulose extinction (Figure 6c). Twins inplagioclase grains are offset across grain-scale faults andindividual twins vary in thickness along their length, mostprominently in bent crystals. Elongate and rounded por-phyroclasts of clinozoisite, feldspar, and occasionallyamphibole are commonly fractured (see, for example,Figures 6d, 6e, and 6f ). The fractures usually show littleto no displacement and some fractures contain quartz,chlorite, or epidote. Continuous grain fracturing and theseparation of broken fragments along the foliation appear tobe the principal process resulting in grain size reduction inthese samples. Fractured fragments still in contact with oneanother are angular. Awide range of grain sizes is present inthese samples that commonly contain bands of ultramylon-ite up to 1 mm thick.[41] East of the westernmost thrust in the imbricate zone

(e.g., top of Figure 6d), narrow bands (up to 5 cm thick) ofultramylonitic S2ITF folia that display the same kinematicsas the thrust zones are common. Rocks from these areascontain a bimodal grain size with small porphyroclasts ofclinozoisite and less commonly feldspar widely dispersedthrough a very fine-grained matrix of quartz, feldspar,chlorite, and epidote that are elongate in the foliation.Feldspar porphyroclasts show well-developed asymmetrictails of recrystallized matrix material. The long axes ofinequant grains have variable orientations relative to themylonitic foliation and many porphyroclasts have smoothoutlines with no obvious mantle. The well-rounded por-phyroclasts show very little evidence of internal deforma-tion other than minor fracturing. This characteristic brittle tosemi-brittle failure of feldspars in thrust zones more than500 m from the Western Fiordland Orthogneiss contactcoincides with the appearance of chlorite as a main folia-tion-defining metamorphic mineral. These features, whichaccompany a change from granulite to amphibolite faciesconditions, imply deformation temperatures of <550�C orslightly higher strain rates [Simpson, 1985; Pryer, 1993].Temperatures estimated using microstructures indicate rela-tively cool conditions for development of S2CG outside theimbricate thrust zone and are slightly lower than the 550�–600�C calculated using thermobarometric methods.[42] In summary, we have found that adjacent to and

within the Western Fiordland Orthogneiss, feldspar in thrustzones was dynamically recrystallized, but in thrust zonesmore than 500 m from the contact, feldspar displayed semi-brittle behavior. The results of microstructural analysiscompare reasonably well with our calculated thermometry.However, the thermometric estimates of temperature in thehigh-grade thrust zones have large errors (2s error bars onthese results were of the order of 100�–150�C), and wethink that the results are potentially high (T = 750�–800�C).Our microstructural analyses confirm that the temperatureof deformation was >550�C. There are four observationsthat suggest the temperature of deformation was probably<750�–800�C. (1) At temperatures >700�C, we wouldexpect to see evidence of rapid recovery of quartz thatwould result in strain-free grains as outlined by Passchier

Figure 6. (continued)


and Trouw [1995]. This was not a microstructural feature ofthe high-grade thrust zones. (2) There was very limitedevidence of secondary grain growth in quartz because ofgrain boundary area reduction as would be expected attemperatures >700�C [Olsen and Kohlstedt, 1985]. (3)Flame albite in K-feldspar was common in the WesternFiordland Orthogneiss thrust zones. At temperatures>700�C, this would not be expected [Pryer, 1993]. (4)Myrmekite is abundant at temperatures >700�C [Simpson,1985]. We observed only a few small myrmekite grains.For these reasons we feel that 650�–700�C is a morereasonable temperature range for the D2ITF deformationwithin and near the contact zone of the Western FiordlandOrthogneiss. A second interpretation is that deformation inthe thrust zones continued during cooling of the batholith.High-grade metamorphic minerals that grew and equili-brated while the batholith was still very hot may have beenpreserved during later thrusting when the microstructureswe observed developed.

6.3. Relationships Between Deformation andMagmatism

[43] We have shown a clear link between metamorphicconditions, feldspar microstructure, and proximity to theWestern Fiordland Orthogneiss during D2ITF deformation.We suggest that because the Western Fiordland Orthog-neiss and the adjacent country rock were still relativelyhot (650�–700�C or possibly higher) at the time ofdeformation, imbricate thrust faulting must have followedshortly after emplacement of the batholith in the CaswellSound region. This is supported by the similar conditionspreserved in the George Sound diatexites and the imbri-cate thrust zone at Caswell Sound. Pressure estimates ofP = 7–9 kbar from the imbricate thrust zone indicate adepth of approximately 25–30 km for thrust faulting inthe Caswell Sound region. Metamorphic data from theGeorge Sound (P = 7–9 kbar) diatexites suggest anemplacement depth of �25–30 km for the WesternFiordland Orthogneiss batholith.

7. Sense of Shear Indicators

[44] In this section we describe microstructural andmacroscopic indicators used to determine the senses ofdisplacement during formation of the Caswell fold-thrustbelt. Boudinaged grains and layering show that the L2CG

and L2ITF mineral lineations are true stretching directions.In our analysis, we compare microstructures and lineationdirections in the zone of imbricate thrust faults with those inthe zone of polyphase folding and ductile shear.[45] Fault propagation folds occur at the termination

points of many ultramylonitic thrust ramps (e.g., Figure 3b).These structures indicate west up over east senses ofdisplacement (or Caswell Gneiss over Western FiordlandOrthogneiss). Other macroscopic indicators of sense ofdisplacement include �0.5 m wide recumbent overfoldsof sedimentary or igneous layers with inverted limbs thatare often sheared out by minor thrust zones displaying west

up over east offsets (Figure 4c). In the high-grade thrustzones within and proximal to the Western Fiordland Orthog-neiss, shear zone cleavages (C0 cleavage and S-C fabric inFigure 6d), and >90% of asymmetric tails on feldspar augenand asymmetric mica fish indicate west up over east senseof shear. Distal to the Western Fiordland Orthogneiss,antithetic (Figure 6b) and synthetic (Figure 6c) microfaultsin feldspar grains and asymmetric quartz microstructures(e.g., oblique shape-preferred foliation) also indicate westup over east in the imbricate thrust zone. Surfaces orientedparallel to the L2ITF lineations and perpendicular to S2ITF

foliation planes record the maximum amount of asymmetryindicating that the lineations coincide with the direction oftectonic transport. The consistency of kinematic indicatorsand lineation directions within the zone of imbricate thrustzones also suggest that these structures are reliable indica-tors of displacement sense and direction, respectively.[46] L2CG lineation measurements in the zone of poly-

phase folding and ductile shear plunge gently to the southand south-southeast at high angles (�60�–90�) to L2ITF inthe zone of imbricate thrust faults. However, unlike thezone of imbricate thrusts, asymmetric structures and well-developed sense of shear indicators do not occur withinthe zone of polyphase folding and ductile shear. In thenarrow west vergent thrust domain asymmetric structuresrecord east over west displacements parallel to a moder-ately east plunging stretching lineation. If D2CG and D2ITF

deformation were simultaneous or progressive across thefold-thrust belt, then the differences in L2CG and L2ITF

lineation orientations indicate that north-south stretching inthe zone of polyphase folding and ductile shear occurredduring east directed thrust transport in the zone of imbri-cate thrust faults.

8. Kinematic Model of Fold-Thrust Belt

Evolution

[47] The two-sided geometry of the Caswell fold-thrustbelt defined by the opposing orientations of west vergentthrusts and east vergent imbricate thrusts suggests that thesethrust zones resulted from subhorizontal (east-west) com-pression and may represent conjugate structures. The kine-matics, structural style, and metamorphic assemblageswithin the two sets of thrusts are consistent with themrepresenting conjugates. Figure 7a represents an interpreta-tion of structures and intrusions as they may have lookedprior to the development of the Caswell fold-thrust belt andshortly after emplacement of the Western Fiordland Orthog-neiss. Our analyses indicate that rheologically weak zonespreferentially localized contractional deformation into thrustplanes. This latter result is supported by the preferentialoccurrence of thrust zones in the marble layers of theCaswell Gneiss, in the zone of thermally weakened crustin the Western Fiordland Orthogneiss aureole, and betweenthe marbles and the undeformed McKerr monzodiorite in thewest vergent thrust zone (Figure 7b). The geometry ofupright F2CG folds and the steep orientation of S2CG inthe zone of open folding also are consistent with east-westshortening and subhorizontal compression.


Figure 7. (a) Summary diagram of the intrusive relationships of the Western Fiordland Orthogneissbatholith prior to development of the Caswell fold-thrust belt. The Western Fiordland Orthogneiss isinferred to have been emplaced as a series of shallowly dipping sills that most commonly have shallowlydipping contacts with the country rocks [Bradshaw, 1990]. The lines labeled S1

WFO represent a shallowlydipping foliation (possible of magmatic origin) within the Western Fiordland Orthogneiss. Themetasedimentary units are migmatized within the contact aureole of the Western Fiordland Orthogneiss.(b) P-T path followed by the Western Fiordland Orthogneiss and adjacent country rock. The WesternFiordland Orthogneiss was loaded soon after emplacement and therefore would have followed thevertical full line shown in P-T space. The Paleozoic country rocks were much cooler and would havebeen loaded to similar depths and heated slightly. There was probably not enough time for them to heatsubstantially [Bradshaw, 1989a]. The Paleozoic country rocks within the aureole would have heated upbefore or during loading (curved full line). (c) Interpretative cross section of the Caswell fold-thrust beltshowing basal decollement at the roof of the Western Fiordland Orthogneiss. (d) Regional section acrossthe Median Batholith and Eastern and Western Provinces of New Zealand showing the relative position(box) of the Caswell fold-thrust belt at the root of the Early Cretaceous magmatic arc. Note two-sidedthrust belt at top of the Western Fiordland Orthogneiss (WFO). Evidence of convergence below the fold-thrust belt is from Daczko et al. [2001a].


[48] In the zone of open folding the subhorizontal S1CG-L1CG fabric forms the dominant structure. Thrust surfaces areabsent in this zone, and F2CG folds are broad open warps ofS1CG. S2CG is widely spaced and weakly developed andrecrystallization of the S1CG-L1CG fabric is minimal. Theseobservations suggest that the zone of open folding is an areaof low D2CG strain. We have confirmed this interpretationusing the simple method of measuring the final versus initiallengths of folded S1CG surfaces in parts of this zone toestimate the amount of D2CG shortening parallel to thedominant transport direction of the fold-thrust belt. Thesemeasurements may vary with different folding mechanismsbut suggest that as little as 5% shortening by F2CG foldingoccurred in most parts of the zone of upright folding.[49] In the zones of polyphase folding and imbricate

thrusts F2CG folds are tight to isoclinal and minor thrustzones shear out F2CG fold limbs. Linear strain measure-ments using folded foliation surfaces suggest that as muchas �75% shortening by F2CG folding occurred during D2CG

deformation in these two areas. In accordance with theestimates of linear strains, the S2CG foliation in these areasis intensely developed and commonly transposes the olderS1CG foliation. These observations confirm that the zonesof polyphase folding and imbricate thrust faults are areasof high D2CG/D2ITF strain. East of the imbricate thrust zoneand west of the west vergent thrust zone, the effects ofcontractional deformation are minimal to nil (Figure 2c).[50] We have shown that strain is localized at the contact

between the Western Fiordland Orthogneiss and CaswellGneiss. This information combined with the listric geometryof the imbricate thrusts suggests that a buried decollementsurface should occur at depth (Figure 7b). The WesternFiordland Orthogneiss outcrops to the north and south ofCaswell Sound (George Sound and Charles Sound, respec-tively; Figure 1), and hence this unit is inferred to be belowthe exposed level of the Caswell Gneiss and Caswell fold-thrust belt at Caswell Sound. The geometry of open folding,the lack of thrust zones, and the small degree of shortening(�5%) in the zone of upright folding located structurallyabove the Western Fiordland Orthogneiss contact alsosuggest that the Caswell Gneiss is detached from the West-ern Fiordland Orthogneiss. The original intrusive geometryof the Western Fiordland Orthogneiss may also have con-trolled the geometry of the cross section shown in Figure 7b.For example, Bradshaw [1990] interprets the WesternFiordland Orthogneiss as having been emplaced as a seriesof shallowly dipping sills and this geometry would havebeen more favorable to the development of a buried decolle-ment surface at depth compared with an original steepwalled geometry to the Western Fiordland Orthogneiss.[51] Strain localization in the zone of polyphase folding

and ductile shear and the formation of west vergent thruststhat separate highly deformed marbles and schists from theundeformed McKerr monzodiorite strongly suggests thepresence of a buried thrust ramp. Similar geometries ofconjugate thrusts or backthrusts located above buried rampsinvolving crystalline thrust sheets are well documented infold-thrust belts in other convergent settings [e.g., Klepeis,1994, Figure 3b; McClay and Buchanan, 1992; Merle,

1998, Figure 61]. Delamination of the section along weakmarble horizons between two converging, strong blocksrepresented by the undeformed, unmetamorphosed McKerrmonzodiorite and the crystalline Western Fiordland Orthog-neiss can explain the observed geometry (Figure 7b). Theinferred existence of this buried thrust ramp is consistentwith our observations that thrusts in the imbricate zoneinvolve the Western Fiordland Orthogneiss. We also suggestthat the localized north-south stretching in this zone resultedfrom the ductile extrusion and subsequent spreading ofweak, ductilely deforming paragneisses and marblesbetween these two converging rigid blocks. The observedpatterns resemble other zones of ductile extrusion betweencrystalline thrust sheets such as the Morcles nappe in theHelvetic Alps [Merle, 1989; Ratschbacher et al., 1991].

9. Discussion

[52] In this section we discuss our results in context withpublished models of the development of the Fiordland high-pressure granulites. We also evaluate the significance of ourresearch for interpreting processes affecting the deep levelsof magmatic arcs.[53] On the basis of regional correlations and geochro-

nological work completed south of Caswell Sound [e.g.,Ireland and Gibson, 1998] the fabric we have labeledS1CG is most likely Paleozoic in age. The age of S1WFO,however, must be Early Cretaceous on the basis of well-documented 126–116 Ma ages from the Western Fiord-land Orthogneiss [Mattinson et al., 1986; Gibson et al.,1988; Gibson and Ireland, 1995; Muir et al., 1998]. HenceS1CG and S1WFO are unrelated. Both these foliations areoverprinted by D2 deformation and all thrust zones of theCaswell fold-thrust belt, indicating an Early Cretaceousage for the development of the fold-thrust belt.[54] Metamorphic assemblages within the contact aureole

between Paleozoic country rock and the Western FiordlandOrthogneiss at George and Caswell Sounds indicate anemplacement depth of �25–30 km (P = 7–9 kbar) for thepluton. The contact at Caswell Sound is deformed by a seriesof imbricate east vergent thrust zones that formed underhigh-grade (granulite and amphibolite facies) metamorphicconditions and form part of an �2-km-wide, two-sided fold-thrust belt (Figures 2 and 7). The Western Fiordland Orthog-neiss is cut by garnet granulite reaction zones in whichenstatite and hornblende are mantled by garnet-clinopyrox-ene-bearing assemblages that reflect �25 km of Early Creta-ceous burial [Bradshaw, 1989a; Clarke et al., 2000; Daczkoet al., 2001b]. These structural relationships and metamor-phic assemblages within and adjacent to the pluton implythat the Western Fiordland Orthogneiss underwent tectonicloading by the imbrication of thrust sheets following middleto deep crustal emplacement of the pluton in the EarlyCretaceous. This contractional style of deformation andloading predate the onset of extension at �108 Ma (age ofextension from Tulloch and Kimbrough [1989]).[55] Three models have been proposed previously to

explain the origin of the high-pressure granulite faciesconditions in Fiordland. First, Gibson et al. [1988] and


Gibson and Ireland [1995] suggested that the WesternFiordland Orthogneiss intruded into the lower crust (at P >12 kbar) within a divergent tectonic regime. They inferredthat ductile normal faulting subsequently tectonicallyexhumed the Western Fiordland Orthogneiss during themid-Cretaceous (�105 Ma). Second, Mattinson et al.[1986], Bradshaw [1989a], and Bradshaw and Kimbrough[1989] hypothesized a middle to deep crustal (P = 6–7 kbar)emplacement depth for the Western Fiordland Orthogneissand used tectonic loading by overthrusts to explain a Creta-ceous up-pressure history recorded in western Fiordland.They observed metamorphic assemblages interpreted torepresent P = 6–7 kbar that were overprinted by high-pressure assemblages at P = 12–13 kbar in the Paleozoicparagneiss. Jadeite zoning patterns in clinopyroxene andgarnet-clinopyroxene granulite assemblages that cut two-pyroxene hornblende granulite facies assemblages in theWestern Fiordland Orthogneiss also supported their model.Finally, Oliver [1990] and Brown [1996] proposed a modelinvolving magma loading to explain the observations madeby Bradshaw [1989a]. This model was attractive because atthat time no thrust sheets had been discovered in westernFiordland to support tectonic loading by overthrusts.SHRIMP dating of monazite in Paleozoic paragneiss atDoubtful Sound suggested a Paleozoic age of the assemb-lages used by Bradshaw [1989a] to infer a Cretaceous up-pressure metamorphic history [Ireland and Gibson, 1998],throwing doubt on models involving Cretaceous loading.However, Clarke et al. [2000] used thermodynamic model-ing of Cretaceous metamorphic assemblage changes in theArthur River Complex and Western Fiordland Orthogneissin northern Fiordland (Figure 1) to confirm a Cretaceous up-pressure loading of these rocks. Daczko et al. [2001a]presented quantitative kinematic analyses that indicatedintense lower crustal (P = 14 kbar) contraction within a pureshear dominated flow regime and ductile thrust faultingthat followed the Cretaceous loading history reported byClarke et al. [2000]. This deformation occurred at depths of>45 km and possibly represents contractional orogenesis inthe deepest crust that followed the development of theCaswell fold-thrust belt at midcrustal levels. Nevertheless,the data presented by Daczko et al. [2001a], like those wepresent here from Caswell Sound, indicate that shorteningdominated the postemplacement history of the batholith inwestern Fiordland (Figures 7c and 7d).[56] This paper presents the first evidence from western

Fiordland that a fold-thrust belt style of contractionaldeformation deformed the upper contact of the Early Creta-ceous Western Fiordland Orthogneiss. This result supportsan interpretation of the tectonic loading of both the WesternFiordland Orthogneiss and its country rocks to explain theup-pressure history recorded by metamorphic assemblages.However, we point out that this result does not exclude acomponent of magma loading during emplacement of theFiordland high-pressure granulite facies belt. Nevertheless,it does indicate that a compressional tectonic regimeaccompanied and outlasted emplacement of the voluminousWestern Fiordland Orthogneiss [cf. Gibson et al., 1988;Gibson and Ireland, 1995; Ireland and Gibson, 1998].

[57] Studies of other magmatic belts in North America,South America, and elsewhere have shown that the thick-ness of arc crust generally increases through time [Cowardet al., 1986; Brown and Walker, 1993; Tobisch et al., 1995;Crawford et al., 1999; Miller et al., 2000]. One questionthat is relevant to all arc systems is how this thickeningoccurs and whether large contractional faults are involved inthe process. For example, in the northern Cascades of thenorthwest United States, an up-pressure metamorphic his-tory is recorded in changing metamorphic mineral assemb-lages at the roots of a Cretaceous arc, but the driving forceof burial and whether it involved thrust faulting, magmaticloading, pure shear thickening, or some other mechanism iscontroversial (see Miller et al. [2000] for a review). In thecentral Coast Mountains Batholith of northern BritishColumbia and southeast Alaska, a midcrustal contractionalhistory is implied by regional relationships and the kine-matics of penetrative ductile fabrics [Klepeis et al., 1998;Crawford et al., 1999]. However, evidence of discrete thrustzones and the effects of possible thrusting on the mechan-ical and thermal evolution of this arc mostly has beenobliterated by the formation of migmatites, pluton emplace-ment, and postcontractional ductile normal faulting. TheFiordland data are important to the study of magmatic arcsbecause they indicate that loading via the imbrication ofthrust sheets in the deep crust is a viable mechanism for thecrustal thickening observed in arc environments. Theserelationships also provide direct evidence that thrust loadingcontributed to the characteristic up-pressure pattern ofmetamorphism which has been observed in other magmaticbelts in different settings [e.g., Brown and Walker, 1993;Paterson and Miller, 1998; Whitney et al., 1999].[58] Some studies of magmatic belts also suggest that

pluton emplacement mechanisms and processes controllingthe evolution of arc crust vary with depth. For example,Miller and Paterson [1999] proposed that plutons in someparts of the northern Cascades representative of upper crustascended as viscoelastic diapirs enclosed by host rocks thatwere deforming by both ductile and brittle processes. Buoy-ancy rather than regional stress was inferred to control theascent of magma in this setting. In contrast, studies of deepcrustal exposures have shown that displacements alongductile faults also can control the emplacement of sheet-like intrusive bodies [Klepeis and Crawford, 1999; Craw-ford et al., 1999]. The Fiordland data support interpretationsthat large vertical displacements at the deepest levels ofmagmatic arcs may commonly involve tectonic imbricationand the development of discrete thrust zones.[59] Finally, we suggest that the partitioning of thrust

zones into thermally softened crust at Caswell Soundsuggests that the emplacement geometry and orientationof the contacts of the Western Fiordland Orthogneissinfluenced the style of contractional deformation affectingthe batholith. At Caswell Sound, thrust zones reactivated thegently dipping Western Fiordland Orthogneiss/country rockcontact and the gently dipping S1CG layering in countryrocks. In contrast, the steep-walled geometry of the WesternFiordland Orthogneiss contact at George Sound appears tohave shielded the contact at that locality from developing


fold-thrust belt geometries. These data support interpreta-tions of deformation patterns in other crustal sections [e.g.,Miller and Paterson, 2001] by suggesting that mechanicalanisotropies created by inherited compositional layeringinfluenced the style of contractional deformation in thedeep crustal portions of the batholith in Fiordland.

10. Conclusions

[60] Structural and metamorphic data from the middle-lower crustal roots of a composite batholith in westernFiordland show that loading by the imbrication of thrustsheets in the deep crust is a viable mechanism for thecrustal thickening in arc environments. We documentimbricate, granulite facies to amphibolite facies thrustzones that form part of a newly discovered middle-lowercrustal (25–30 km) fold-thrust belt that cuts the uppermostcontact of the Early Cretaceous (126–116 Ma) WesternFiordland Orthogneiss. Our data indicate that contractionoutlasted emplacement of the Western Fiordland Orthog-neiss and controlled the tectonic evolution of the batholith.Thrust faulting produced mylonitic to ultramylonitic fab-rics in a well-defined �3-km-wide zone of imbricate thrustfaults that formed within and at the margin of the WesternFiordland Orthogneiss. These thrust zones contain high-

grade garnet-biotite-K-feldspar granulite facies assemb-lages within 500 m of the contact and chlorite-epidoteamphibolite facies assemblages further from the WesternFiordland Orthogneiss contact. The localization of high-grade mylonitic to ultramylonitic thrust zones at and nearthe Western Fiordland Orthogneiss contact implies thatthrust zones were preferentially partitioned into crust thatwas thermally weakened by magmatism. Our results showthat large vertical displacements within an Early Creta-ceous magmatic arc were linked to thrust imbrication inthe deep crust and produced an up-pressure metamorphichistory that is similar to that observed in other magmaticbelts worldwide.

[61] Acknowledgments. Funding to support this work was providedby two University of Sydney Institutional Australian Research Councilgrants to K. A. Klepeis; a large Australian Research Council grant to K. A.Klepeis and G. L. Clarke (A10009053), and a National Science Foundation(EAR-0087323) grant to K. A. Klepeis and T. Rushmer. An Australianpostgraduate award from the University of Sydney supported N. R. Daczkoduring preparation of this manuscript. We are especially grateful to NickMortimer and Andy Tulloch of the IGNS, Dunedin for many helpfuldiscussions and logistical assistance, and to the Department of Conservationin Te Anau for permission to visit and sample localities in the FiordlandNational Park. We also thank N. M. Kelly for assistance in the field andlaboratory and Gabriela Mora-Klepeis for many helpful discussions.

References

Andronicos, C. L., L. S. Hollister, C. Davidson, andD. Chardon, Kinematics and tectonic significanceof transpressive structures within the Coast Pluto-nic Complex, British Columbia, J. Struct. Geol.,21, 229– 243, 1999.

Bishop, D. G., J. D. Bradshaw, and C. A. Landis,Provisional terrain map of South Island, NewZealand, in Tectonostratigraphic Terranes of the

Circum-Pacific Region, edited by D. G. Howellet al., pp. 512 – 522, Circum-Pacific Counc. forEnergy and Resour., Houston, Tex., 1985.

Blattner, P., The North Fiordland transcurrent conver-gence, N. Z. J. Geol. Geophys., 34, 543 – 553,1991.

Bradshaw, J. D., Cretaceous geotectonic patterns in theNew Zealand region, Tectonics, 8, 803 –820, 1989.

Bradshaw, J. D., A review of the Median TectonicZone: Terrane boundaries and terrane amalgama-tion near the Median Tectonic Line, N. Z. J. Geol.Geophys., 36, 117 –125, 1993.

Bradshaw, J. Y., and D. L. Kimbrough, Comment: Ageconstraints on metamorphism and the developmentof a metamorphic core complex in Fiordland,southern New Zealand, Geology, 17, 380 – 381,1989.

Bradshaw, J. Y., Origin and metamorphic history of anEarly Cretaceous polybaric granulite terrain, Fiord-land, southwest New Zealand, Contrib. Mineral.

Petrol., 103, 346 –360, 1989a.Bradshaw, J. Y., Early Cretaceous vein-related garnet

granulite in Fiordland, southwest New Zealand: Acase for infiltration of mantle-derived CO2-richfluids, J. Geol., 97, 697–717, 1989b.

Bradshaw, J. Y., Geology of crystalline rocks ofnorthern Fiordland: Details of the granulite faciesWestern Fiordland Orthogneiss and associatedrock units, N. Z. J. Geol. Geophys., 33, 465 –484, 1990.

Brown, E. H., High-pressure metamorphism caused bymagma loading in Fiordland, New Zealand, J. Me-

tamorph. Geol., 14, 441 –452, 1996.

Brown, E. H., and N. W. Walker, A magma-loadingmodel for Barrovian metamorphism in the south-east Coast Plutonic Complex, British Columbia andWashington, Geol. Soc. Am.. Bull., 105, 479–500,1993.

Carter, R. M., C. A. Landis, R. J. Norris, and D .G.Bishop, Suggestions towards a high-level nomen-clature for New Zealand rocks, J. R. Soc. N. Z., 4,29 –84, 1974.

Clarke, G. L., K. A. Klepeis, and N. R. Daczko,Cretaceous high-P granulites at Milford Sound,New Zealand: Their metamorphic history andemplacement in a convergent margin setting,J. Metamorph. Geol., 18, 359 –374, 2000.

Coward, M. P., B. F. Windley, R. D. Broughton, I. W.Luff, M. G. Petterson, C. J. Pudsey, D. C. Rex, andM. Asif Khan, Collision tectonics in the northwestHimalayas, in Collision Tectonics, edited by M. P.Coward and A. C. Ries, Geol. Soc. Spec. Publ., 19,203 –219, 1986.

Crawford, M. L., K. A. Klepeis, G. Gehrels, andC. Isachsen, Batholith emplacement at mid-crustallevels and its exhumation within an obliquely con-vergent margin, Tectonophysics, 312, 57 – 78,1999.

Daczko, N. R., K. A. Klepeis, and G. L. Clarke, Evi-dence of Early Cretaceous collisional-style orogen-esis in northern Fiordland, New Zealand and itseffects on the evolution of the lower crust, J. Struct.Geol., 23, 693 –713, 2001a.

Daczko, N. R., G. L. Clarke, and K. A. Klepeis, Trans-formation of two-pyroxene hornblende granulite togarnet granulite involving simultaneous meltingand fracturing of the lower crust, Fiordland, NewZealand, J. Metamorph. Geol., 19, 549 – 562,2001b.

Ferry, J. M., and F. S. Spear, Experimental calibrationof the partitioning of Fe and Mg between biotiteand garnet, Contrib. Mineral. Petrol., 66, 113 –117, 1978.

Gibson, G. M., and T. R. Ireland, Granulite formation

during continental extension in Fiordland, Nature,375, 479– 482, 1995.

Gibson, G. M., and T. R. Ireland, Extension of Dela-marian (Ross) orogen into western New Zealand:Evidence from zircon ages and implications forcrustal growth along the Pacific margin of Gond-wana, Geology, 24, 1087–1090, 1996.

Gibson, G. M., I. McDougall, and T. R. Ireland, Ageconstraints on metamorphism and the developmentof a metamorphic core complex in Fiordland,southern New Zealand, Geology, 16, 405 – 408,1988.

Graham, C. M., and R. Powell, A garnet-hornblendegeothermometer: Calibration, testing, and applica-tion to the Pelona Schist, southern California,J. Metamorph. Geol., 2, 13 –31, 1984.

Grocott, J., M. Brown, R. D. Dallmeyer, G. K. Taylor,and P. J. Treloar, Mechanism of continentalgrowth in extensional arcs: An example from theAndean plate boundary zone, Geology, 22, 91 –394, 1994.

Hill, E. J., A deep crustal shear zone exposed in wes-tern Fiordland, New Zealand, Tectonics, 14, 1172–1181, 1995a.

Hill, E. J., The Anita Shear Zone: A major, middleCretaceous tectonic boundary in northwesternFiordland, N. Z. J. Geol. Geophys., 38, 93 – 103,1995b.

Hodges, K. V., and F. S. Spear, Geothermometry,geobarometry and the Al2SiO5 triple point atMt. Moosilauke, New Hampshire, Am. Mineral.,66, 1118–1134, 1982.

Holland, T. J. B., and R. Powell, An enlarged andupdated internally consistent thermodynamic dataset with uncertainties and correlations: The systemK2O-Na2O-CaO-MgO-MnO-FeO-Fe2O3-Al2O3-TiO2-SiO2-C-H2O-O2, J. Metamorph. Geol., 8,89 –124, 1990.

Ingram, G. M., and D. H. W. Hutton, The Great Tona-lite sill: Emplacement into a contractional shearzone and implications for Late Cretaceous to early


Eocene tectonics in southeastern Alaska and BritishColumbia, Geol. Soc. of Am. Bull., 106, 715–728,1994.

Ireland, T. R., and G. M. Gibson, SHRIMP monaziteand zircon geochronology of high-grade meta-morphism in New Zealand, J. Metamorph. Geol.,16, 149–167, 1998.

Kimbrough, D. L., A. J. Tulloch, E. Geary, D. S.Coombs, and C. A. Landis, Isotope ages from theNelson region of South Island, New Zealand:Structure and definition of the Median TectonicZone, Tectonophysics, 225, 433–448, 1993.

Kimbrough, D. L., A. J. Tulloch, D. S. Coombs, C. A.Landis, M. R. Johnston, and J. L. Mattinson, Ura-nium-lead zircon ages from the Median TectonicZone, New Zealand, N. Z. J. Geol. Geophys., 37,393–419, 1994.

Klepeis, K., Relationship between uplift of the meta-morphic core of the southernmost Andes and short-ening in the Magallanes foreland fold and thrustbelt, Tierra Del Fuego, Chile, Tectonics, 13, 882–904, 1994.

Klepeis, K. A., and M. L. Crawford, High-temperaturearc-parallel normal faulting and transtension at theroots of an obliquely convergent orogen, Geology,27, 7 – 10, 1999.

Klepeis, K. A., M. L. Crawford, and G. Gehrels, Struc-tural history of the crustal-scale Coast shear zonenear Portland Canal, Coast Mountains orogen,southeast Alaska and British Columbia, J. Struct.Geol., 20, 883–904, 1998.

Klepeis, K. A., N. R. Daczko, and G. L. Clarke, Kine-matic vorticity and the tectonic significance ofsuperposed mylonites in a major lower crustal shearzone, northern Fiordland, New Zealand, J. Struct.Geol., 21, 1385–1405, 1999.

Landis, C. A., and D. S. Coombs, Metamorphic beltsand orogenesis in southern New Zealand, Tectono-physics, 4, 501 –518, 1967.

MacKinnon, T. C., Origin of the Torlesse terrane andcoeval rocks, South Island, New Zealand, Geol.Soc. Am. Bull., 94, 967–985, 1983.

Mattinson, J. L., D. L. Kimbrough, and J. Y. Bradshaw,Western Fiordland orthogneiss: Early Cretaceousarc magmatism and granulite facies metamorphism,New Zealand, Contrib. Mineral. Petrol., 92, 383 –392, 1986.

McClay, K. R., and P. G. Buchanan, Thrust faults ininverted extensional basins, in Thrust Tectonics,edited by K. R. McClay, pp. 93 – 104, Chapmanand Hall, New York, 1992.

Merle, O., Strain patterns within spreading nappes,Tectonophysics, 165, 57 –71, 1989.

Merle, O., Emplacement Mechanisms of Nappes andThrust Sheets, 159 pp., Kluwer Acad., Norwell,Mass., 1998.

Miller, R. B., and S. R. Paterson, In defense of magmaticdiapirs, J. Struct. Geol., 21, 1161–1173, 1999.

Miller, R. B., and S. R. Paterson, Influence of litholo-gical heterogeneity, mechanical anisotropy, andmagmatism on the rheology of an arc, North Cas-cades, Washington, Tectonophysics, 342, 351–370,2001.

Miller, R. B., E. H. Brown, D. P. McShane, and D. L.Whitney, Intra-arc crustal loading and its tectonicimplications, North Cascades crystalline core, Wa-shington and British Colombia, Geology, 21, 255 –258, 1993.

Miller, R. B., S. R. Paterson, S. M. DeBari, and D. L.Whitney, North Cascades Cretaceous crustal sec-tion: Changing kinematics, rheology, metamorph-ism, pluton emplacement and petrogenesis from 0to 40 km depth, in Guidebook for Geological Field

Trips in Southwestern British Columbia and North-ern Washington, edited by G. L. Woodsworth et al.,pp. 229 – 278, Geol. Assoc. of Can., CordilleranSect., St. John’s, NF, Canada, 2000.

Mortimer, N., Triassic to Early Cretaceous tectonicevolution of New Zealand terranes: A summaryof recent data and an integrated model, in Proceed-

ings of the 1995 PACRIM Congress, AustralasianInstitute of Mining and Metallurgy, Carlton, Victor-

ia, Australia, edited by J. L. Mauk and J. D. St.George, pp. 401–406, Australas. Inst. of Min. andMetal., Parkville, Vic., Australia, 1995.

Mortimer, N., A. J. Tulloch, R. N. Spark, N. W. Walker,E. Ladley, A. Allibone, and D. L. Kimbrough,Overview of the Median Batholith, New Zealand:A new interpretation of the geology of the MedianTectonic Zone and adjacent rocks, J. Afr. Earth Sci.,29, 257 –268, 1999.

Muir, R. J., S. D. Weaver, J. D. Bradshaw, G. N. Eby,and J. A. Evans, The Cretaceous Separation Pointbatholith, New Zealand: Granitoid magmas formedby melting of a mafic lithosphere, J. Geol. Soc.London, 152, 689–701, 1995.

Muir, R. J., S. D. Weaver, J. D. Bradshaw, G. N. Eby,J. A. Evans, and T. R. Ireland, Geochemistry ofthe Karamea Batholith, New Zealand, and com-parisons with the Lachlan Fold Belt granites ofSE Australia, Lithos, 39, 1 – 20, 1996.

Muir, R. J., T. R. Ireland, S. D. Weaver, J. D. Brad-shaw, J. A. Evans, G. N. Eby, and D. Shelly,Geochronology and geochemistry of a Mesozoicmagmatic arc system, Fiordland, New Zealand,J. Geol. Soc. London, 155, 1037– 1053, 1998.

Oliver, G. J. H., An exposed cross-section of continen-tal crust, Doubtful Sound Fiordland, New Zealand:Geophysical and geological setting, in Exposed

Cross – Sections of the Continental Crust, editedby M. H. Fountain and D. M. Salisbury, pp. 43 –69, Kluwer Acad., Norwell, Mass., 1990.

Oliver, G. J. H., and J. H. Coggon, Crustal structure ofFiordland, New Zealand, Tectonophysics, 54, 253 –292, 1979.

Olsen, T. S., and D. L. Kohlstedt, Natural deformation

and recrystallisation of some intermediate plagio-clase feldspars, Tectonophysics, 111, 107 – 131,1985.

Passchier, C. W., and R. A. J. Trouw, Microtectonics,Springer-Verlag, New York, 1995.

Paterson, S. R., and R. B. Miller, Magma emplacementduring arc-perpendicular shortening: An examplefrom the Cascades crystalline core, Washington,Tectonics, 17, 571 –586, 1998.

Powell, R., and T. J. B. Holland, An internally consis-tent data set with uncertainties and correlations, 3,Applications to geobarometry, worked examplesand a computer program, J. Metamorph. Geol., 6,173 –204, 1988.

Pryer, L. L., Microstructures in feldspars from a majorcrustal thrust zone: The Grenville Front, Ontario,Canada, J. Struct. Geol., 15, 21 –36, 1993.

Ratschbacher, L., H. R. Wenk, and M. Sintubin, Cal-cite textures: Example from nappes with strainpath partitioning, J. Struct. Geol., 13, 369– 384,1991.

Simpson, C., Deformation of granitic rocks across thebrittle-ductile transition, J. Struct. Geol., 7, 503–511, 1985.

Suppe, J., Principals of Structural Geology. 537 pp.,Prentice Hall, Old Tappan, N. J., 1985.

Tobisch, O. T., J. B. Saleeby, P. R. Renne, B. McNulty,and T. Weixing, Variations in deformation fieldsduring development of a large-volume magmaticarc, central Sierra Nevada, California, Geol. Soc.Am. Bull., 107, 148 –166, 1995.

Wandres, A. M., S. D. Weaver, D. Shelley, and J. D.Bradshaw, Change from calc-alkaline to adakiticmagmatism recorded in the Early Cretaceous Dar-ran Complex, Fiordland, New Zealand, N. Z. J.Geol. Geophys., 41, 1 – 14, 1998.

Whitney, D. L., R. B. Miller, and S. R. Paterson, P-T-tconstraints on mechanisms of vertical tectonic mo-tion in a contractional orogen, J. Metamorph.

Geol., 17, 75 –90, 1999.Williams, J. G., and C. T. Harper, Age and status of the

Mackay Intrusives in the Eglinton-Upper Holly-ford area, N. Z. J. Geol. Geophys., 21, 733– 742,1978.

Wood, B. L., Metamorphosed ultramafites and asso-ciated formations near Milford Sound, NewZealand, N. Z. J. Geol. Geophys., 15, 88– 127,1972.

��G. L. Clarke and N. R. Daczko, School of

Geosciences, University of Sydney, Building F05,NSW 2006, Australia. ([email protected];[email protected])

K. A. Klepeis, Department of Geology, Universityof Vermont, Burlington, VT 05405-0122, USA.([email protected])


Date post:	28-Aug-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Thermomechanical evolution of the crust during convergence...

Documents